International banking and large community banks: a preliminary look.

Jesswein, Kurt R.

INTRODUCTION

Banking has had a rich and illustrious history. Much of its development can be directly related to the banks' ability to provide trade financing, credit, and foreign exchange services, activities specifically designed to overcome many of the impediments to conducting trade and investment transactions that cross political and economic borders.

This paper provides an introductory look at the trends and current state of the U.S. banking industry, specifically those institutions often referred to as large community banks, with regards to providing banking products that service international business and commerce. Despite the reputed benefits to a bank's bottom line (directly from generating fee and interest income, and indirectly from more fully developing customer relationships), the trend has been decidedly negative, as more and more institutions appear to be abandoning international financing during a time period of increasing globalization in business.

BRIEF HISTORY OF INTERNATIONAL BANKING

From Mesopotamia in the third millennium B.C., to Babylon, Greece, Rome, and even to the money changers described in the Bible, elements of banking enterprise have long been a part of human development. "Modern" banking can probably be most directly traced to the twelfth and thirteenth centuries with the rise of great Italian merchant families such as the Medici who began providing banking services essential to the development of international commerce.

Most notable to the advancement of modern banking was the introduction of bills of exchange, financial instruments that provided a means of exchanging currency across Europe. They also provided an important means of providing short-term credit (given the inevitable delays in transporting documents and currency), with interest indirectly charged through the rate of exchange and thus avoiding the prohibitions against usury (Green, 1989).

The next great advances came with the rise of the first truly international merchant banking houses, most notably the Baring Brothers and the Rothschild family, through whom banking, and in particular international banking, first began to reach the Americas. For example, Barings provided the financing for the United States government's purchase of Louisiana from the French and the Rothschilds are reputed to have been involved in financing both sides of the American Civil War. Later, banking houses began to arise within the United States itself, led by names such as Morgan, Oppenheim, and Drexel, alongside the continued development of international banks in Europe and Asia. (Green, 1989)

Although two World Wars and a global depression tempered some of the periodic booms in international banking and finance through the first half of the twentieth century, the Bretton Woods and GATT (General Agreement on Tariffs and Trade) accords after the Second World War provided stability in international monetary relationships and liberalization of trade and capital movements. International banking thrived as the global economy exploded under this new found liberalization.

Concurrently, the introduction and development of the Eurodollar market provided banks the means to overcome local and international restrictions on their international operations, and led to an explosion in institutions providing international banking services. The breakdown of Bretton Woods in the early 1970s, which provided the impetus to floating exchange rates, coupled with the recycling of petrodollars through the Euromarkets beginning with the first OPEC (Organization of Petroleum Exporting Countries) oil embargoes of the mid 1970s, provided institutions with the instability of markets and the ready supply of funds to introduce a wide range of international banking products and services.

It is also in this time frame that we find some of the earliest calls for expanding the banking activities by U.S. institutions into the international realm. It was argued that even smaller, regional banks should move into the international arena as a means of retaining their competitive stance, expanding their reach, and ultimately improving their profits (Hardy, 1972; Thoman, 1975). Yet, as one examines the extent of involvement in providing various international services, one finds a surprising lack of such involvement, other than by the largest of the money center banks. And despite the anecdotal references to the expected profitability of offering international services (e.g., providing credit, trade financing, or foreign currency services), it appears that the number of banks offering such services has actually dwindled over time.

In addition, despite the readiness of a select group of smaller banks to follow the suggestions of developing international services, there has been a dearth of studies examining the success or failure of institutions offering international banking services. A key exception has been the work by Haslem, et al, who, in a series of papers (1986, 1992, 1995), looked at the relationship of various international banking strategies and bank profitability; however, these papers only focused on larger, money center banks.

On the other hand, one area that has been examined, even for smaller banks, is the use of financial derivatives. Given the high growth of the derivatives markets in the late 1980s and into the 1990s and beyond, there has been much work done in examining bank usage of derivative products, both from a trading and from a risk management perspective. Brewer, et al (1996), Carter and Sinkey (1998), and Sinkey and Carter (2000) have all looked at how derivatives use was related to overall bank performance or how it impacted bank hedging decisions.

The focus of this paper is to examine the scope of international banking activities provided by banks, particularly larger community banks within the United States, and to begin to look at the apparent aversion to providing such services over recent years. We look at both the depth and breadth of international banking among U.S. banks, looking for factors that determine whether or not a bank provides such services, and to what extent.

DATA SOURCES AND SAMPLE SELECTION

In developing this study, we begin by defining what constitutes international activities within the scope of banking operations. Probably the broadest yet most practical definition is one that includes those banks that provide some sort of international activity that can be documented through the data provided in their Call Reports delivered to the Federal Deposit Insurance Corporation (FDIC). All balance sheet, income statement, and international activity data used in this study are collected from these Call Reports, and these datasets are available for download from the Federal Reserve Bank of Chicago (www.chicagofed.org). A distinguishing feature of this paper is the use of quarterly rather than annual data due to the short-term nature of most international financing transactions, particularly letter of credit and foreign currency transactions.

The specific sample of banks examined in this paper is often identified in the literature as large community banks. These are typically defined as institutions having total assets between $100 million and $1 billion (Carter & Sinkey, 1998; Gilbert & Sierra, 2003; DeYoung, Hunter & Udell, 2004). However, to capture a small subset of institutions not often examined, we also include banks that are referred to as mid-sized banks (Ennis, 2004) and thus also have banks with total assets up to $10 billion. Smaller banks, those below $100 million in total assets, are excluded from the study due to the assumption that they are typically too small or too localized to have significant demands placed on them to provide international services, or that they would not have sufficient resources to devote to such endeavors. The largest banks (above $10 billion in total assets) are excluded because they are often involved with other aspects of international banking (e.g., derivatives) that may be more closely related to their trading or hedging functions and not necessarily for the purpose of meeting customer demands.

Reviewing the call report data over the past four years (2002--2005), we find that approximately one percent of banks in the database had asset sizes greater than $10 billion and approximately forty-nine percent had asset sizes less than $100 million. This results in us having around fifty percent of all banks providing call reports being defined as large community banks.

To allow for the comparability of results across time and eliminate problems associated with the steadily growing size of banks affecting the definition of large community bank, we have arbitrarily chosen to exclude the upper one percent and the lower forty-nine percent of banks for each period studied. For example, during the third quarter of 2005, there were a total of 7,985 banks available for the study, but 3,913 banks (49 percent) were eliminated from the bottom and 80 banks (1 percent) were eliminated from the top, leaving a sample size of 3,993 banks. During the first quarter of 1990, the first period included in the analysis, 13,024 banks were available for the study, but 6,382 banks (49 percent) were eliminated from the bottom and 130 banks (1 percent) were eliminated from the top, leaving a sample size of 6,512 banks.

In addition, we also specifically examined one subset of banks, namely those that provided letters of credit. Other individual subsets were excluded because the number of banks uniquely providing other types of international activities (direct lending or foreign exchange) is relatively small in comparison. For example, during the third quarter of 2005, of the 7,985 banks in the overall sample, 1,263 (15.8 percent) provided some sort of international activity. And of those 1,263 banks, over 84 percent (1,064) only provided letters of credit, while less than 5 percent (60) only provided loans, and around 1 percent (15) only provided foreign exchange. Comparable figures exist for the sample of large community banks as almost 82 percent (597) of the total of 731 banks providing international services only provided letters of credit. In fact, earlier periods show an even greater domination of letters of credit, with nearly 90 percent of all banks (and 88 percent of community banks) providing letters of credit as their only form of international exposure during the initial period of this study, namely the first quarter of 1990.

METHODOLOGY AND VARIABLE SPECIFICATION

A variety of analyses are performed. We focused on three items: letters of credit, direct foreign lending, and foreign exchange. Letters of credit (LTRCRDTS) are measured as the nominal value of all commercial letters of credit outstanding, as standardized by the bank's total assets. Letters of credit are arguably the simplest way and certainly the most common way in which banks provide international banking services. Two other variables (INTLOANS and FOREX) measure the nominal value of the bank's international commercial loans and outstanding foreign currency positions, respectively. A generic international banking exposure variable (TOTINTLX) is also derived as the sum of the outstanding balances of the three different types of international services. Furthermore, to measure whether or not a particular bank provided any type of international services, a dummy variable was created to denote whether or not a particular bank offered any specific international services.

Other operational and financial variables used in the analysis include the following. To capture the impact of firm size, we use the natural log (LNTOTAST) of a bank's total assets. Size is examined due to the tendency for larger banks, even within our sample group, to be more likely involved with providing international services than smaller institutions.

To capture the impact of capital considerations, we include equity capital as a percentage of total assets (EQRAT). Risk-based capital figures may be more appropriate, but such data has only been available since 1996. Thus, for any analyses covering periods since 1996, we also examine the bank's risk-adjusted capital rate (RISKWGT). Equity (or risk-based) capital is included to control for a bank's ability to engage in international activities, much of it off balance sheet, as allowed by regulatory agencies. Institutions with greater amounts of capital would be expected to be more likely to engage in typically riskier international activities.

To measure the profitability of a bank's core operations we used net interest margin (NIMEARN) as measured by net interest income (e.g., total interest income minus total interest expense, divided by total earning assets. Net interest margin is included to control for the relative profitability of the firm's traditional bank activities. Lower profitability is expected to be positively related to the provision of international services as less-profitable institutions may look towards such services as means of generating additional profits.

We also examine various accounting performance figures. Although likely strongly related to the net interest margin, these standard accounting ratios, such as return on assets or ROA, measured by net income divided by total assets, and return on equity or ROE, as measured as net income divided by total equity, were included to capture the profitability of the entire bank's activities, including its non-interest sources of revenue and operating expenses.

To examine short-term investing risks, we include rate sensitive assets (RSA), defined as investment assets (mainly securities and loans) set to mature within one year or that carry variable interest rates. We also examine short-term financing risks by including rate sensitive liabilities (RSL), defined as financing or deposit liabilities set to mature within one year or that carry variable interest rates. We also include the relationship between a bank's rate sensitive assets and rate sensitive liabilities, commonly referred to as its interest rate gap (GAP), which is measured as the nominal difference between RSA and RSL. All three variables (RSA, RSL, and GAP) were standardized by a bank's total assets. Because long foreign currency and letter of credit positions potentially impact the overall rate-sensitivity of an institution by creating more rate-sensitive assets, we include these variables to examine the impact of an institution's existing interest rate exposure on its involvement in international activities.

Besides looking at the overall sensitivity of a bank's source of funds captured by rate sensitive liabilities, we also examine its deposit structure by looking at its reliance on interest-bearing deposit liabilities (INTDEP) and non-interest bearing deposit liabilities (NOINTDEP). We also look at the relative use of interest-bearing deposits over non-interest bearing deposits (INTDEPCT). Reliance on relatively more-expensive and perhaps less stable interest-bearing deposits could preclude a bank from engaging in international activities that may further tighten its overall liquidity or harm its profitability.

Lastly, because the vast majority of international activities are credit-based, we examined the bank's existing lending position. We include loans as a percentage of total assets (LOANS) to examine the impact of a bank's relative focus on lending. Banks more focused on the lending function domestically would be assumed to be more likely to engage in international credit activities. However, those banks with potential credit problems, measured by allowance for loan losses as a percentage of total loans (ALLPCT), would be less likely to engage in international credit activities that, because of their greater risks, could compound the bank's credit problems.

Parametric and nonparametric tests were conducted to look for significant differences between those banks providing international banking services and those that do not. In addition, logistic regressions are used to simultaneously examine those factors most related to banks providing international services. The model is adapted from one used by Carter and Sinkey (1998) in their examination of derivative use by banks. The logit models examine the differences between those banks providing international services and those that did not as represented by a dummy variable. Variables examined included bank size (LNTOTAST), capital (EQRAT), the profitability of operations (NIMEARN), rate-sensitivity of assets (RSA) and of liabilities (RSL), reliance on interest-bearing deposits (INTDEPCT), credit-focus of bank (LOANS), and extent of credit portfolio weakness (ALLPCT). Minor variations of the model included replacing the EQRAT variable with the more appropriate risk-based (RISKWGT) capital measure for time periods since 1996, replacing the narrow NIMEARN measure with the broader ROA or ROE variables, replacing RSA and RSL with GAP (the difference between RSA and RSL), and examining the use of interest-bearing deposits (INTDEP) or non-interest-bearing deposits (NOINTDEP) individually rather than simultaneously.

EMPIRICAL RESULTS

No matter how one chooses to examine the data, the U.S. banking community has steadily become less involved in international banking activities over time. In 1990, 27 percent of all banks and 33 percent of the community banks provided some type of international banking services, mainly letters of credit. By 2005, however, this number has dropped to 16 and 18 percent, respectively. Measuring the level of international activity as the amount of letters of credit outstanding as a percentage of total bank assets, we find similar results as this value has fallen by two-thirds, from letters of credit equaling 1.2% of all bank assets in 1990, and only 0.4% in 2005. A summary of this trend can be seen graphically in the graph showing Letters of Credit as Percentage of Total Assets presented earlier. Thus, both the number of banks involved with providing international services and the extent of that involvement has been dropping.

We next look for any discernable differences in operating or financial characteristics of those institutions providing international services and those that do not. To get a sense of the direction and current status of international banking within the sample of community banks, the data here are presented at three distinct points in time: the third quarter of 2005 (the most recent data available), and the third quarters of 1998 and 1991, respectively. These points were chosen to give a broad picture of the data across time (note that these results are indicative of other nearby periods of time). Similar points in time within each year (third quarter) were chosen to avoid problems associated with the apparent seasonality of the data. Although some rigor is lost by not performing more stringent time series analysis on the relationships, we instead choose to concentrate on a snapshot approach with which one can make generalizations.

As neither the assumption of homogeneity of variances nor the normality of the data is consistent across the variables used in this study (as discerned by Brown & Forsythe and Kolmogorov-Smirnov tests on both the general sample and the sample of community banks), both parametric (paired sample t-tests) and nonparametric (Wilcoxon) procedures were run. The tables below summarize the mean values of each variable comparing providers and non-providers of international services at the designated three points in time. Satterthwaite t-values and Wilcoxon Z-scores are provided to show any significant differences between the two groups.

A review of the tables shows both a significant similarity of results across time and in some cases, noteworthy differences. First, even within the subset of banks examined (large community banks), size certainly matters as larger institutions are much more likely to be involved with providing international banking services. Bank profitability, as measured by return on assets, also appears to be a factor as banks lacking in profits may be drawn to the potential revenues and profits from international banking. However, when we examine profitability as measured by net interest margin, we find a distinct difference between the results from 2005 and 1998 with those from 1991. When looking at the data from 1991, we arrive at the opposite conclusion; that is, it appears that net interest margin and international banking were positively related. This difference may reflect the significant change in banking fortunes as the early 1990s was a far less profitable time period for banks as compared with more recent periods.

When we look at the risk characteristics of bank operations, we find that the use of interest-bearing deposits as a source of funds appears to be a significant impediment for engaging in international banking activities. Banks relying on less costly, and perhaps more stable, non-interest bearing deposits tend to be more often drawn to international banking. Similarly, the extent of capitalization (whether defined as equity or in risk-based terms) appears to not be the factor assumed as more heavily capitalized institutions are less likely to engage in international banking activities. Thus, relying on more secure deposits may preclude the need for capital for engaging in international banking activities. This may result from the preferential treatment accorded letters of credit (as opposed to more traditional forms of credit) in that capital is only required to match 20 percent of their value vis-a-vis 100 percent for other credits.

Lastly, rather than looking at the alternative variables individually, we instead examine them simultaneously using logistic regressions. The aim is to determine those factors most likely to allow us to differentiate between those institutions involved in international banking from those that are not involved. As the results are relatively consistent across time, we focus on the logit model from the most recent (third quarter of 2005) period, the results of which are summarized in Table 5.

The results indicate that besides bank size, banks more heavily invested in short-term and other rate-sensitive assets were surprisingly more likely to add to this amount by engaging in international banking activities. Banks with lower levels of outstanding loans were also much more likely to engage in international activities. In addition, and confirming the earlier discussion, banks with more stable non-interest bearing deposit liabilities and those with lesser amounts of risk-based capital were also apparently drawn to international banking.

CONCLUSIONS AND DISCUSSION OF FUTURE RESEARCH

Whether or not a bank engages in international banking is a complex issue. Although there are potential gains to the bottom-line from generating additional revenues and from further developing customer relationships, there may be serious impediments, both apparent and imagined, to providing international services. Large community banks in the U.S. have been moving away from the international arena and the reasons why are varied and complex.

We have presented an initial glimpse into some of the complexities associated with this topic. The dominant (and relatively obvious) association of bank size and international banking activities indicates a need to more closely examine other potential factors besides the size of the institution. Furthermore, more complex time series analysis of the changing nature of the industry may shed light on how some of the apparent relationships have changed or are changing with an attempt then to perhaps forecast the future direction of the industry. In addition, more thorough analyses of the various relationships among the variables will be able to add depth to our understanding of the topic.

Given that international competitiveness in the commercial and industrial sectors are of such vital importance to the long-term strength of the U.S. economy, and that the importance of the financing of those commercial activities can not be overstated, a better understanding of the issues facing commercial banks can have wide ranging implications.

REFERENCES

Brewer III, E., W.E. Jackson III & J.T. Moser (1996). Alligators in the swamp: the impact of derivatives on the financial performance of depository institutions. Journal of Money, Credit & Banking, 28(3), 482-497.

Carter, D.A, & J.F. Sinkey, Jr. (1998). The use of interest rate derivatives by end-users: the case of large community banks. Journal of Financial Services Research 14(1), 17-34.

DeYoung, R, W.C. Hunter, & G.F. Udell (2004). The past, present, and probable future for community banks. Journal of Financial Services Research, 25(2/3), 85-133.

Ennis, H.M. (2004). Some recent trends in commercial banking. Economic Quarterly (Federal Reserve Bank of Richmond), 90(2), 41-61.

Gilbert, R.A. & G.E. Sierra (2003). The financial condition of U.S. banks: how different are community banks. Review (Federal Reserve Bank of St. Louis), 85(1), 43-56.

Green, E. (1989). Banking: an illustrated history. New York: Rizzoli International.

Hardy, C.C. (1972). International banking: small- and medium-sized inland banks told they can profit by filling need. Banking, 65(1), 19-23.

Haslem, J.A., A. Christofi, J.P. Bedingfield & A.J. Stagliano (1986). A statistical analysis of international banking measures and relative profitability. Management International Review, 26(2), 5-13.

Haslem, J.A., C.A. Scheraga & J.P. Bedingfield (1992). An analysis of the foreign and domestic balance sheet strategies of the U.S. banks and their association to profitability performance. Management International Review, 32(1), 55-75.

Haslem, J.A., C.A. Scheraga & J.P. Bedingfield (1995). An analysis of the impact of international activity on the domestic balance sheet of U.S. banks. Management International Review, 35(1), 45-68.

Sinkey, J.F., Jr. & D.A. Carter (2000). Evidence of the financial characteristics of banks that do and do not use derivatives. Quarterly Review of Economics and Finance (40), 431-449.

Thoman, G.R. (1975). International banking can be profitable for U.S. regional banks. Columbia Journal of World Business, 10(4), 23-32.

Kurt R. Jesswein, Sam Houston State University

Table 1 Sample breakdown

Period 2005 2004 2003 2002 2002-05

Average number
 of banks 7998 8140 8303 8446 8224
Number with assets 3653 3877 4126 4446 4032
< $100 million (45.7%) (47.6%) (49.7%) (52.6%) (49.0%)
Number with assets 89 91 90 86 89
> $10 billion (1.1%) (1.1%) (1.1%) (1.0%) (1.1%)

Source: FDIC Call Reports

Table 2 Differences in Means between Providers and Non-providers
(Third Quarter: 2005)

 Providers Non- t-value
 providers

LNTOTAST 13.1250 12.5650 11.87
ROA 0.0116 0.0122 -1.91
ROE 0.1223 0.1237 -0.48
NIMEARN 0.0298 0.0309 -3.73
EQRAT 0.0998 0.1027 -1.90
RISKWGT 0.1450 0.1561 -3.37
RSA 0.2648 0.2646 0.02
RSL 0.3932 0.4035 -1.92
GAP -0.1280 -0.1390 1.25
INTDEP 0.6561 0.6768 -4.41
NOINTDEP 0.1370 0.1315 1.62
INTDEPCT 0.8267 0.8372 -2.42
LOANS 0.6605 0.6746 -2.25
ALLPCT 0.0125 0.0131 -1.79

 Pr > [absolute Z-score Pr > [absolute
 value of t] value of Z]

LNTOTAST <.0001 10.55 <.0001
ROA 0.0562 -0.50 0.6150
ROE 0.6334 0.44 0.6624
NIMEARN 0.0002 -2.48 0.0129
EQRAT 0.0576 -1.49 0.1370
RISKWGT 0.0008 -3.24 0.0012
RSA 0.9864 -0.39 0.6958
RSL 0.0550 -1.92 0.0550
GAP 0.2105 0.66 0.5088
INTDEP <.0001 -4.80 <.0001
NOINTDEP 0.1048 1.16 0.2462
INTDEPCT 0.0158 -2.27 0.0228
LOANS 0.0248 -2.58 0.0097
ALLPCT 0.0736 0.41 0.6841

Table 3 Differences in Means between Providers and Non-providers
(Third Quarter: 1998)

 Providers Non- t-value
 providers

LNTOTAST 12.6230 12.0830 13.70
ROA 0.0120 0.0126 -1.72
ROE 0.1260 0.1265 -0.18
NIMEARN 0.0321 0.0330 -2.91
EQRAT 0.1001 0.1022 -1.47
RISKWGT 0.1617 0.1712 -3.14
RSA 0.2817 0.2655 3.18
RSL 0.4474 0.4599 -2.90
GAP -0.1660 -0.1940 4.28
INTDEP 0.6963 0.7138 -4.52
NOINTDEP 0.1274 0.1193 3.31
INTDEPCT 0.8428 0.8534 -3.09
LOANS 0.6148 0.6171 -0.48
ALLPCT 0.0150 0.0145 1.50

 Pr > [absolute Z-score Pr > [absolute
 value of t] value of Z]

LNTOTAST <.0001 11.96 <.0001
ROA 0.0854 -2.43 0.0151
ROE 0.8567 -0.13 0.8940
NIMEARN 0.0037 -1.50 0.1331
EQRAT 0.1420 -3.33 0.0009
RISKWGT 0.0017 -5.28 <.0001
RSA 0.0015 2.97 0.0030
RSL 0.0038 -2.70 0.0068
GAP <.0001 4.11 <.0001
INTDEP <.0001 -5.88 <.0001
NOINTDEP 0.0010 3.12 0.0018
INTDEPCT 0.0020 -3.90 <.0001
LOANS 0.6307 -1.03 0.3016
ALLPCT 0.1336 4.61 <.0001

Table 4 Differences in Means between Providers and Non-providers
(Third Quarter: 1991)

 Providers Non- t-value
 providers

LNTOTAST 12.1220 11.6920 15.06
ROA 0.0070 0.0091 -1.65
ROE 0.0608 0.0849 -2.17
NIMEARN 0.0324 0.0320 2.18
EQRAT 0.0818 0.0845 -3.09
RISKWGT *
RSA 0.4890 0.4524 8.03
RSL 0.4674 0.4685 -0.30
GAP 0.0217 -0.0160 7.89
INTDEP 0.7488 0.7701 -8.05
NOINTDEP 0.1269 0.1090 10.06
INTDEPCT 0.8537 0.8742 -9.12
LOANS 0.5743 0.5578 4.00
ALLPCT 0.0187 0.0172 3.22

 Pr > [absolute Z-score Pr > [absolute
 value of t] value of Z]

LNTOTAST <.0001 11.93 <.0001
ROA 0.0984 -2.28 0.0225
ROE 0.0301 -0.54 0.5883
NIMEARN 0.0296 5.12 <.0001
EQRAT 0.0020 -5.35 <.0001
RISKWGT *
RSA <.0001 6.76 <.0001
RSL 0.7623 -4.10 <.0001
GAP <.0001 7.74 <.0001
INTDEP <.0001 -11.11 <.0001
NOINTDEP <.0001 10.53 <.0001
INTDEPCT <.0001 -11.03 <.0001
LOANS <.0001 4.39 <.0001
ALLPCT 0.0013 4.30 <.0001

* Risk-weighted capital was first reported in 1996

Table 5 Logit Model Results (Third Quarter: 2005)

 Maximum
 Likelihood Wald
 Estimate Odds Ratio Chi-Square Pr > Chi-Sq

Intercept -9.3886 47.7601 <.0001
LNTOTAST 0.6232 1.865 157.8231 <.0001
ROA -1.2039 0.300 0.0129 0.9096
ROE -0.9493 0.387 0.5768 0.4476
NIMEARN -6.7265 0.001 1.1218 0.2895
EQRAT 0.3378 1.402 0.0340 0.8537
RISKWGT -1.9879 0.137 3.4858 0.0619
RSA 0.5877 1.800 3.7636 0.0524
RSL 0.4369 1.548 1.1377 0.2861
GAP *
INTDEP -0.1731 0.841 0.0793 0.7782
NOINTDEP 2.4407 11.481 3.5263 0.0604
INTDEPCT 0.648 1.912 0.3692 0.5434
LOANS -0.7942 0.452 4.7830 0.0287
ALLPCT -3.4554 0.032 0.2511 0.6163

* GAP is not examined due to its direct relationship to the RSA
and RSL variables.